Video Hummus

Yes, add in foliage that is often in these actions shots and it turns into a complete muddy mess. I also think over sharpening contributes to the crappy codec encoding quality.

The 360 Max is the right direction for action cams.

Yes , that company is called FujiFilm. You might have heard of them. Let me remind you it has also taken a decade for CMOS sensors to be where they are at today.

Thanks for pointing this out kye. I haven’t often thought of this advantage. I think it’s also one of the reasons having exceptional AF system on full frame and medium format are even more important. I’m talking about the S lenses. Watch this interview with Matt from Panasonic. At 54:00 minutes in Matt addresses how they are making the new S lenses. https://youtu.be/lQ2cuFfcWB8?t=54m The pro lenses Matt specifically says they are not parfocal but have parfocal abilities because they are doing some focusing tricks with the focusing elements, even in manually focus, to achieve parfocal behavior (timestamp at 56:00) A true optically parfocal lens would be huge. Like the fujinon cine lenses.

It seems most of the add-on accessories are.

I still like the form factor of my Mavic Air better. Would love to see a 1” sensor in a future air form factor (or smaller). Ill trade less flight time for it too.

I agree. It appears they are incorporating a few features in all the new S lenses. Manual focus clutch with a linear focus choice and parfocal abilities by using the AF system to “fake” an optically parfocal lens. I get it. But OIS in a wide lens didn’t really make sense. I suspect adding OIS to the 10-25 just made it too bulky, especially when the GH5/G9/EM1.2 cameras have such good IBIS already. Bottom line is if DoF is on the top of your list then MFT is not your best option. For me, MFT makes all the right trade offs for what I need and want. And as camera phones have shown the future is in computational photography. I’m sure there will be a time where I can shoot on a MFT camera and select a few extra stops of DoF in camera with a push of a button.

No, the 50-200 is harder to manually focus for me. I really wish Panasonic would add linear focus by wire support like the S1. The 10-25 has much better control with the focus clutch for whatever reason. Well Panasonic has shown it can do IBIS with dual gain sensor tech in a full frame. So, they would be smart to try for that in the GH6. I love my GH5S. If I had an option to get GH5S with IBIS as a second camera...I'd be all over it.

Laws of physics. Bits have to be represented by some physicality, smallest you can go is an atom maybe and electron. Problem is, the smaller you go the less those things behave in predictably manners. Quantum physics and all that jazz. Same problem exists with fuel and space travel. Moores Law has essential hit a wall (in the CPU domain) storage isn’t far behind. I know hard drive size has not shrunk in anyway for 10 years or more. That’s a problem. It makes me wonder the immense amount of work YouTube has to do to process and store video. ?

Well, it’ll cost you more than $1799 to pickup a S1H and an L mount lens. I have the 10-25 f1.7. It’s a brilliant lens. It will be even more brilliant on a future GH6. If the GH6 has as good lowlight ability as the GH5S then it’s going to be a huge hit as long as it carries on the GH5 tradition. I shot some video of some Elk a couple of days ago. It was dusk, approaching blue hour. I had my 50-200 f2.8-4. I was shooting at 200mm f4 at ISO 10000 and 12800 in 10bit HLG. The noise performance was more than acceptable. I lost a bit of color vitality but it was dusk so it really didn’t matter. The 4K footage downsampled to 1080p didn’t need any noise reduction at all. If I was to keep it at 4K I would have applied noise reduction to some of the clips at 12800. The images were more than useable. I switched out to the 10-25 for some environment shots at at f1.7 I could drop my ISO to 2500-3200 which is a piece of cake for the GH5S.

I’m not discounting the march of technology. But there are physical limits to storage technology. Storage of any kind takes space. Will 8K be mainstream someday. Yep. Will it be sooner than we think? I’m throwing my hat in the “not as soon as you think” rink.

Matt should think about doing more of these Q&A sessions. He doesn’t do much marketing speech and gives clear answers. Seems very welcoming to forwarding requests to engineers. We will get better cameras this way.

6K is the middle ground as a capture resolution. It makes 4K edits flexible in post while also being able to upscale to 8K without much problems at all. I think 8K is still entirely to heavy for almost anybody except huge production houses like BBC or National Geographic that archive high resolution footage for tv shows and documentaries. 8K will be marketed heavily to the CONSUMER in the next 5 years. Especially following the 2020 Olympics. Im not sure it’s very practical for 95% of other video professionals. If you shoot weddings with hours of footage, would you want to shoot in 8K? The data requirements would be enormous. Probably talking tens of terabytes. Youtubers? No way.

It’s worth noting that Kraig was using the sigma 14-24 f2.8 lens. There would be wobble on the wide end for sure!

KnightsFan is exactly right. it would be like ford patenting a car that is 4 feet or greater in length with a attached roof. Cars existed before the model T but didn’t have a roof. Is a roof not obvious enough to grant ford an patent to a car, which is a highly complex piece of equipment, because they slapped on an obvious, preexisting, roof on it and made it longer than 4 feet? A roof is obvious. The length of said object is obvious. The size of RAW images is obvious. RED didn’t invent bigger images. They didn’t make an invention related to the CMOS sensor. They simply bought exclusive rights to a sensor off the shelf. Slapped on JPEG2000 wavelet compression on RAW images (existed before) and submitted a patent for it. Compressed TIFFs existed before RED “invented” RAW compression. They used them to “demo” their REDCODE. The video shows they didn’t put in 12 months of research. They took pre-existing technologies, with pre-existing ideas and then claimed they were novel. The premise behind REDS patent is that their incredibly novel idea was to capture 2K+ RAW images on a camera and compress the RAW images ON THE CAMERA inline and store it (instead of having a computer do it) Fucking ground breaking! The patent is entirely without merit.

You have lots of options for affordable cameras that shoot good. No one cares about red fanboys here. We care about if RED's patent is valid or not. In this case, details matter.

OK, RED's PR is pretty freaking scummy. Claiming everything was in-house and directly innovated by them when it really wasn't and lying to the press about it is shitty business ethics. However, I still find admiration for the fact that RED basically LEGO built their camera and marketed better than any of the established players. The other companies just moved too slow. Can't wait for part 5 about REDCODE since a lot hinges on that patent. I gotta say this has been a great "investigative" documentary series on YouTube.

The bar for Canon is so low for them to have a hot mirrorless camera. 24p, FF or nearly FF 4K, DPAF in 1080p120, Clog, RF mount. 4Kp60 would be nice but not essential. 10bit in at least the 24p mode. They can be forgiven for only 8bit in higher FPS since Canon colors are great SOOC. Thats it. Add a tilt screen and they will win over most people. I would buy that camera and I don’t own any RF or EF glass. But they won’t do it because they want you to drop 5K plus on the body alone. And here comes Sony, slowing eating their lunch while it’s still in Canons hands with better and better hybrid AF and tracking technology. Pretty soon, Canons largest advantage and biggest feature, having DPAF, won’t mean shit.

Make some popcorn before you watch. Because they seem to be developing it from scratch.

Well, just barely. It's 7 stops in body and with a OIS lens its 7.5 stops. Still pretty amazing.

You should make living quarters in there. Start a YouTube channel called cargo truck life. Title of the first video should be: WHY Cargo TRUCK life DESTROYS Van Life”. You will need to photoshop bikini girl on top of your truck for the clicks maybe have the truck crushing a van with the front tire. You know it’s gold.

I wonder what the DR numbers are for high frame double exposure HDR video.

Agreed. EIS has its limits. I think it’s best used like in the newer Olympus cameras in conjunction with IBIS to mitigate or eliminate that jello effect in the corners using sensor data from the actually IBIS mechanism and not by analyzing the image.

I agree. This thing with a good implementation of double exposure HDR video would be hard to beat as a pocket adventure cam for 99.9% of people. I have yet to see low light electronic stabilization from a phone that doesn’t shit the bed.

Sigma can’t go wrong as long as they thoughtful consider user feedback. The NVMe suggestion should be taken seriously. It makes perfect sense for the form factor and the cDNG uncompressed format they are offering. It’s clear they want RAW in this camera and they can only do uncompressed at the moment. Which means high bitrates and large files sizes. NVMe sticks tick all those boxes plus form factor and price per GB. I can see why Panasonic went with dual SD card slots on the S1H. Why didn’t sigma just go with XQD or Express and offer 12bit uncompressed internal?